<class 'pandas.core.frame.DataFrame'>
Index: 3533892 entries, 0 to 4456613
Data columns (total 25 columns):
 #   Column                            Dtype   
---  ------                            -----   
 0   race                              object  
 1   sex                               object  
 2   co_applicant                      object  
 3   age                               object  
 4   income                            float64 
 5   loan_amount                       int64   
 6   property_value_ratio              float64 
 7   mortgage_term                     object  
 8   credit_model                      object  
 9   debt_to_income_ratio              object  
 10  combined_loan_to_value_ratio      float64 
 11  main_underwriter                  object  
 12  tract_to_metro_income_percentage  object  
 13  lender_type                       object  
 14  lender_size                       int64   
 15  white_population_pct              float64 
 16  metro_name                        object  
 17  metro_code                        object  
 18  metro_size_percentile             object  
 19  state_code                        object  
 20  county_code                       object  
 21  census_tract                      object  
 22  loan_outcome                      object  
 23  lender_id                         category
 24  fips                              object  
dtypes: category(1), float64(4), int64(2), object(18)
memory usage: 680.9+ MB
None

HMDA Dataset Info¶

Mortgage outcomes in HMDA data¶

Applicant Characteristics¶

What leads to loan denials?¶

Are disparities focused in any geographic areas?¶

Denial rates in the Chicago metropolitan area:¶